Piotr Orzechowski, CEO at Infermedica explains why AI provides so many advantages over traditional rule-based decision trees and how integrating AI technology with doctors expertise can enhance patient care.
Anxiety about the introduction of automation in the workplace has been well documented. With some estimations stating that, by 2025, machines will be doing half of all work tasks, perhaps concerns are not entirely misplaced.
In healthcare, robotics and machine learning are already having an impact on patient care, providing basic assistance across many clinical facilities – no doubt a life-saving job. Some may feel that eventually technology – specifically Artificial Intelligence (AI) – will progress to a point where it can deliver advanced consultation, without the need for a physician. For others, interaction with a machine to determine a course of treatment will never be able to replicate the personalised consultation with a physician, the latter of which has an unparalleled level of expertise and training.
Really, however, AI shouldn’t be viewed as an existential threat to doctors but instead a companion. The two can work harmoniously alongside one another, with AI enhancing knowledge, expertise, and time management. As healthcare organisations across the globe embrace digital technologies to tackle the pandemic, harnessing AI capabilities has the potential to greatly improve patient care.
Adoption of technology in healthcare
It’s no secret that technology already has a huge presence in healthcare. From wearable devices sharing information on vitals to secure messaging apps for medical professionals, tech is transforming almost every aspect of patient care. In Q3 of 2020, European start-ups raised €1.6 billion alone, revealing the phenomenal growth of healthtech services.
In 2020, however, the pandemic has shined an unmissable spotlight on the relationship between technology and healthcare. Earlier this year, when healthcare organisations around the world had to cancel routine and non-urgent appointments in order to cope with the demands of the pandemic, the adoption of telemedicine services and e-consultations soared. Virtual appointment services are expected to fundamentally disrupt the healthcare market with telemedicine services expected to grow sevenfold by 2025.
Necessity prevented patients from physically seeing their doctor for minor health complaints, but it brought digitalisation to the forefront of healthcare services around the world. These have been typically slow to innovate, shackled by levels of bureaucracy and complicated procurement processes, which prevented fast innovation and built a barrier to adopting technologies that can improve patient care.
Irreversible trends
And it’s not just telemedicine that has benefited: collaboration between health professionals and technology companies have introduced new solutions and devices in response to the virus at an unprecedented rate. For example, 3D printed valves and big data solutions that provide patient monitoring in real-time are just two introductions that reflect the huge transformation which has taken place.
The healthcare industry has witnessed how collaboration between innovative companies, healthcare systems and doctors can find new ways to improve patient care and respond to the challenges of the pandemic, and it’s inconceivable to imagine that the change which has taken place will be reversed.
In a few months, we’ve seen systems adopt technology rapidly – and physicians incorporating AI systems into their traditional practices is a next step.
Benefits of AI
Healthcare systems are commonly overstretched. Emergency departments and doctors’ offices typically deal with patients who have conditions that could be treated elsewhere. This means doctors and other healthcare workers have more patients to see, which, critically, takes away time that can be spent with patients harbouring more serious issues.
Misdiagnosis is also a problem. By nature, doctors are experts with extensive training, but incorrectly diagnosing patients or missing warning signs can happen occasionally. For example, on average in the UK, it takes around eight years for endometriosis to be correctly diagnosed, with patients commonly prescribed treatment for irritable bowel syndrome (IBS) instead.
AI-driven consultations can alleviate these problems by helping humans, not replacing them. For example, initial consultations conducted by AI solutions can triage patients to the most appropriate level of care, identifying instances where it might be more appropriate for people to seek advice from a pharmacist instead of a doctor. This direction prevents a backlog of patients using the healthcare system unnecessarily, fast-tracking those who need urgent treatment whilst freeing up doctors to spend more time with patients who have complex issues.
Furthermore, AI can provide doctors with enhanced information on a patient’s condition. Systems that analyse health complaints whilst taking into consideration various risk factors, including age, gender and location – as well suggesting possible rarer conditions that share similar symptoms with more common issues, such as endometriosis – equip doctors with more actionable information. This can be shared ahead of an in-person consultation, providing the most up-to-date and accurate information on the patient, improving care and diagnosis.
Cutting time needed to cover basic questions makes in-person consultations more productive, efficient and can help to improve the rate of misdiagnosis. Indeed, a recent study of digital health services in Norway found that 72% of the respondents experienced a better follow-up by their GP as a consequence of using e-consultation, while 58% reported improved quality of their treatment. AI can take this further.
Of course, there may be some reluctance to move toward digital consultations. Bedside manner and patient interaction are a large part of visiting a physician. During a face-to-face meeting, a doctor can read non-verbal cues which may indicate how a patient is truly feeling.
AI can make a difference here too, though, with solutions trained to understand the most nuanced movements or expressions; the system then relays the information to the doctor.
The future of healthcare
Ultimately, a healthcare provider introducing an AI to assist patient care is like an accountant using a calculator. Doctors’ skills are supported by reliable, knowledgeable tool that helps them to do their jobs more effectively. What’s more, AI solutions are constantly improving, drawing on data from real-life use cases to assist doctors in making correct and timely diagnoses.
It’s undeniable that the COVID crisis has pioneered the growth of healthcare solutions around the world, what we now encounter is healthcare systems embracing innovation to address the challenge of the pandemic and beyond. Moving forward, it’s critical to continue embracing technology to improve patient care – and AI will have a huge part to play.